The Macroeconomic Effect of AI: Sizing the Software Engineering Channel (NBER) (www.nber.org)

A new working paper from the National Bureau of Economic Research analyzes the macroeconomic impact of artificial intelligence (AI) on software engineering productivity, revealing significant implications for the broader economy. The researchers employed real-time financial market data to estimate how sensitive each firm's stock performance is to an AI-driven market index, particularly focusing on those firms with a higher payroll share in software engineering. Between November 2022 and December 2025, AI is projected to enhance the market's expected value of software engineering productivity by 32.6%, translating to a potential GDP increase of 3.6%. When considering the boost in R&D productivity from improved software engineering, this GDP effect could rise to 6.5%. The findings are particularly significant as they demonstrate that advancements in AI, especially with the rise of sophisticated coding agents, have the potential to more than double their productivity contributions by mid-2026 compared to late 2025. This insight not only underscores the critical role of software engineering in the AI landscape but also highlights the broader economic transformations that may result from ongoing improvements in AI technologies. As the AI/ML community continues to innovate, understanding these macroeconomic effects will be essential for forecasting future trends in productivity and economic growth.